Tweet
Since July 1, 2014 FIDE changed K-factor for U2400 players. Instead of 15, now K=20 for adults and 40 for juniors U18 and below 2300.
For adults, the change is minor. Rating's fluctuation will be slightly bigger. For juniors the issue is completely different.
K=40 means that last month's performance will have significant impact on new rating, sometimes this performance is more important than initial rating. If some junior plays 19 games, the new FIDE rating will reflect his performance only, no matter what was his initial rating.
Here is an example to better understand this point. 3 juniors have initial rating 2000, 2100 and 2200. Each of them plays 19 games. To simplify the calculation, they play only against 2200 FIDE exactly and draw each game.
First player (initial rating 2000). Expected result: 0.24 point/game. 0.26 * 19 * 40 = 197.6 points. The new rating will be 2198.
Second player (2100). Expected result: 0.36 point/game. 0.14 * 19 * 40 = 106 points. The new rating will be 2206.
Third player (2200). Expected result: 0.5 point/game. 0 * 19 * 40 = 0 points. The new rating will be 2200.
So after just 19 games each of them will have around 2200, no matter what was their old rating.
Sure 19 games/month is a lot, but it's a very realistic number for some juniors. In Canada, July probably could give the most chances to play great number of games. World Open and Canadian Open give 18 games. CYCC and CO (very popular combination) give 16, if every game is rated. If junior plays 16 games, the impact on the new rating will be around 85%, and the old rating has just 15%.
At the end, some funny example. One junior has FIDE rating of 2200. He wants to be a 2300 player. With the new K-factor, it's very simple. He doesn't need to hire a good coach, read chess books, prepare to each game, eat and sleep properly during the tournament... Nothing. All he needs is just to make 2 steps.
1. Artificially reduce the rating as low as possible (sandbagging) before he plans to play a lot.
2. Play a lot.
Let say, he dropped to 2000, and after what played 30 games with performance of 2200. (After all he remains the same junior, no better and no worse - 2200 player).
From the first example he gains 0.26 * 30 * 40 = 312 points. The new one will be 2312. Congratulations!
For adults, the change is minor. Rating's fluctuation will be slightly bigger. For juniors the issue is completely different.
K=40 means that last month's performance will have significant impact on new rating, sometimes this performance is more important than initial rating. If some junior plays 19 games, the new FIDE rating will reflect his performance only, no matter what was his initial rating.
Here is an example to better understand this point. 3 juniors have initial rating 2000, 2100 and 2200. Each of them plays 19 games. To simplify the calculation, they play only against 2200 FIDE exactly and draw each game.
First player (initial rating 2000). Expected result: 0.24 point/game. 0.26 * 19 * 40 = 197.6 points. The new rating will be 2198.
Second player (2100). Expected result: 0.36 point/game. 0.14 * 19 * 40 = 106 points. The new rating will be 2206.
Third player (2200). Expected result: 0.5 point/game. 0 * 19 * 40 = 0 points. The new rating will be 2200.
So after just 19 games each of them will have around 2200, no matter what was their old rating.
Sure 19 games/month is a lot, but it's a very realistic number for some juniors. In Canada, July probably could give the most chances to play great number of games. World Open and Canadian Open give 18 games. CYCC and CO (very popular combination) give 16, if every game is rated. If junior plays 16 games, the impact on the new rating will be around 85%, and the old rating has just 15%.
At the end, some funny example. One junior has FIDE rating of 2200. He wants to be a 2300 player. With the new K-factor, it's very simple. He doesn't need to hire a good coach, read chess books, prepare to each game, eat and sleep properly during the tournament... Nothing. All he needs is just to make 2 steps.
1. Artificially reduce the rating as low as possible (sandbagging) before he plans to play a lot.
2. Play a lot.
Let say, he dropped to 2000, and after what played 30 games with performance of 2200. (After all he remains the same junior, no better and no worse - 2200 player).
From the first example he gains 0.26 * 30 * 40 = 312 points. The new one will be 2312. Congratulations!
Comment